Temporal and contemporaneous disaggregation of multiple economic time series

Abstract

Data-based procedure, discrepancy measure, Kalman filter, mean square error, vector autoregressive models, Primary 62M10, secondary 62F30, 62H12,

Ex-post andex-ante prediction of unobserved economic time series: a case study

“…, m(n 1), when the aggregated ®gure Y n1 is available. Guerrero and Nieto (1994) have solved the problem in the multivariate context using state-space models and the Kalman ®lter. Here, we use that approach but with slight modi®cations.…”

“…A NEW SOLUTION TO THE EX-POST ESTIMATION PROBLEM Here we follow some ideas of Guerrero and Nieto's (1994) paper about the multivariate disaggregation case. Let {Z t } be an unobservable stochastic process and assume we observe the process {Y i }, i 1, .…”

Ex post and ex ante prediction of unobserved multivariate time series: a structural‐model based approach

“…It is useful for accomplishing optimal temporal or contemporaneous disaggregation (benchmarking) of multivariate time series, as in the case of both annual/quarterly and sectorial/geographical aggregates. The approach presented here is more flexible and parsimonious than previous results for the same problem, as is the case of, among others, Guerrero and Nieto (1999). This property is due to the use of structural models, in Harvey's (1989) sense.…”

“…This work extends the univariate setting of Nieto (2004) and the multivariate one of Guerrero and Nieto (1999). It is useful for accomplishing optimal temporal or contemporaneous disaggregation (benchmarking) of multivariate time series, as in the case of both annual/quarterly and sectorial/geographical aggregates.…”

“…In the multivariate context, i.e., when the process of interest to be predicted is multivariate, some authors have produced partial results. Among others, Fernandez-Macho et al (1987), Guerrero and Nieto (1999), and the references therein, have worked on the topic.…”